Objectives: Immune thrombocytopenia (ITP) causes increased platelet destruction and suboptimal platelet production, increasing risk of bleeding. This analysis uses a Bayesian metaregression model to indirectly compare effectiveness of the thrombopoietin mimetics romiplostim and eltrombopag for increasing platelet counts, and contrasts the results with those of non-Bayesian approaches.

Methods: Ten databases were searched during 2010. Placebo-controlled trials of 24 weeks’ duration were included. An indirect comparison was undertaken using Bayesian metaregression, which includes all trials in a single model. This was compared with previous analyses in which data for each intervention were combined using simple pooling, logistic regression or meta-analysis, followed by indirect comparison of pooled values using the Bucher method.

Results: Two trials of romiplostim and one of eltrombopag were included. The indirect evidence suggests romiplostim significantly improves overall platelet response compared with eltrombopag. Bayesian metaregression gave an odds ratio (OR) for eltrombopag versus romiplostim of 0.11 (95 percent credible interval 0.02–0.66); p values and Bayesian posterior probabilities ranged from 0.01 to 0.05 for all analyses. There was no significant difference in durable platelet response in any of the analyses, although the direction of effect favored romiplostim (OR = 0.15; 95 percent credible interval, 0.01–1.88); p values and Bayesian posterior probabilities ranged from 0.08 to 0.40 across analyses. Results were relatively consistent between analyses.

Conclusions: Bayesian metaregression generated similar results to other indirect comparison methods, and may be considered the most robust as it incorporates all data in a single model and accounts appropriately for parameter uncertainty.